Fast food box packaging defect detection equipment and system

Abstract

The invention provides a fast food box packaging defect detection device and system, and belongs to the technical field of fast food box detection appliances. The defect identification module is used for receiving the surface image of the fast food box and performing visual identification analysis to obtain a visual analysis result of the fast food box; the visual recognition module is also used for performing visual recognition analysis on the water level and bubbles in the annular water tank in the fast food box defect detection process to obtain an air tightness analysis result; and the defect judgment module is used for comparing the surface defect evaluation score and the air tightness score with a preset threshold value, and marking the fast food box as an unqualified piece if any value is greater than the preset threshold value. The surface defect can be accurately captured from multiple dimensions, the air tightness of the fast food box can be deeply detected, the water level change of the annular water tank and bubble generation feature information are covered, comprehensive analysis from the surface to the air tightness is achieved, the detection process is highly automatic, efficient and rapid, and the large-scale production requirement is met.

Description

Technical Field

[0001] The present invention relates to the technical field of fast food box detection tools, and particularly to a fast food box packaging defect detection device and system. Background Art

[0002] In the current digital and fast-paced social life, the takeaway industry has witnessed unprecedented vigorous development. As a key carrier in the takeaway delivery link, the fast food box plays a crucial role. Whether it is Chinese stir-fried dishes with rice, a variety of covered rice dishes, classic Western hamburger and fries sets, or delicate Japanese sushi, bento and other various delicacies, merchants will pack them into fast food boxes and then quickly deliver them to consumers through takeaway riders. This model has greatly expanded the service scope of the catering industry, fully meeting the needs of people who don't want to go out but desire to taste different delicacies in various scenarios such as busy work or comfortable home life. However, during the extensive application of fast food boxes, quality problems have gradually emerged. Among them, cracks in fast food boxes are one of the most common and influential problems.

[0003] Currently, the methods for detecting whether there are defects in fast food boxes on the market are relatively traditional and single. Generally, the fast food box is filled with water, and whether there is leakage is observed to judge whether there are defects. Therefore, a more efficient, accurate and less impactful detection device and system for fast food boxes are needed to solve the problems encountered above. Summary of the Invention

[0004] Aiming at the deficiencies of the existing technology, the present invention provides a fast food box packaging defect detection device and system to solve the problems raised in the above background art.

[0005] The object of the present invention can be achieved by the following technical solutions: A fast food box packaging defect detection device includes a top cover. The bottom surface of the top cover is fixedly connected with a box body, and a mounting groove is opened in the middle of the top surface of the top cover. A sealing ring is arranged on the inner wall of the mounting groove, and a detection table is installed inside the mounting groove. A circular water groove is opened in the middle of the top surface of the detection table, and a fixing component is installed on the top surface of the detection table.

[0006] Preferably, an installation opening penetrating through to the inner top surface of the box body is opened in the middle of the top surface of the detection table, and an air outlet head is installed on the top surface of the detection table. The lower end of the air outlet head is placed inside the installation opening, and a connecting pipe is sleeved on the lower end of the air outlet head.

[0007] Preferably, an air pump is installed inside the box body. The air outlet of the air pump is connected to the other end of the connecting pipe, and an air inlet is opened in the middle of the rear end surface of the box body. A filter screen is installed inside the air inlet.

[0008] Preferably, connection seats are installed at four ends of the top surface of the detection table. Electric push rods penetrate through the connection seats. A baffle is fixedly connected to the far ends of the four electric push rods, and a rubber plate is fixedly connected to the near ends of the four electric push rods. A spring is arranged between the rubber plate and the connection seat, and the spring is sleeved on the telescopic end of the electric push rod.

[0009] Preferably, a pressure rod is horizontally arranged at the upper end of the top cover. The cross-section of the pressure rod is U-shaped, and mounting rods are vertically and fixedly connected to the rear of both ends of the pressure rod.

[0010] Preferably, movable grooves are formed in the far surfaces of the lower ends of the two mounting rods, and a flipping motor is arranged at the far ends of the two mounting rods. The flipping motor is installed on the top cover. The output end of the flipping motor is connected to one end of a connecting rod arranged in the movable groove. A torsion spring is sleeved on the surface of the connecting rod, and the other end of the torsion spring is fixedly connected to the inner wall of the movable groove. The connecting rod is installed on the flipping motor.

[0011] A fast food box packaging defect detection system includes a vision setting module, an image acquisition module, a defect recognition module, a defect determination module, and a result output module; The vision setting module is used to set a number of high-definition cameras on the detection table, and the high-definition cameras are used to collect the surface images of the fast food boxes; The image acquisition module is used to collect the surface images of the fast food boxes and send them to the defect recognition module; The defect recognition module is used to receive the surface images of the fast food boxes for visual recognition and analysis to obtain the visual analysis results of the fast food boxes; the visual analysis results include the difference in contour area, the difference in contour perimeter, the similarity, the color comprehensive shadow value, the missing shadow value, and the surface defect evaluation score of the surface images; It is also used to perform visual recognition and analysis on the water level and bubbles in the annular water tank during the defect detection of the fast food boxes to obtain the airtightness analysis results; among them, the airtightness analysis results include the statistical index of the water level height, the water level influence value, the standard deviation of any feature in the generation characteristic information of the bubbles and the corresponding statistical index, the standard influence value corresponding to the feature, the bubble influence value, and the airtightness score; The defect determination module is used to compare the surface defect evaluation score and the airtightness score with their corresponding preset thresholds. If any value of the surface defect evaluation score or the airtightness score is greater than its preset threshold, it means that the fast food box has defects, and the fast food box is marked as a non-conforming part; otherwise, if both the surface defect evaluation score and the airtightness score are less than their preset thresholds, it means that the fast food box has no defects, and the fast food box is marked as a conforming part; The result output module is used to receive the visual analysis results and the airtightness analysis results sent by the defect recognition module, and store the visual analysis results and the airtightness analysis results in the database.

[0012] Preferably, the surface image of the fast food box is received for visual recognition analysis, specifically as follows: Obtain the surface image of the fast food box collected by any camera, process the surface image using an edge detection algorithm to obtain the edge contour of the fast food box, calculate the area and perimeter of the edge contour respectively to obtain the contour area value and the contour perimeter value; obtain the standard contour area and the standard contour perimeter of the fast food box collected by this camera; subtract the standard contour area from the contour area value to obtain the contour area difference, and subtract the standard contour perimeter from the contour perimeter value to obtain the contour perimeter difference; set the standard edge contour of the fast food box collected by this camera; use the cosine similarity algorithm to calculate the similarity between the edge contour and the standard edge contour; Enlarge the surface image of the fast food box into a pixel grid image, set the standard color value of the fast food box in the pixel grid image corresponding to the camera, identify the color value of any pixel grid in the pixel grid image, and subtract the standard color value of the pixel grid from the color value of the pixel grid to obtain the standard color difference value of the pixel grid; divide the image within the edge contour of the fast food box into several monitoring areas according to the common shapes, sizes and positions where defects are likely to occur of the fast food box, and assign area influence weights to the monitoring areas; calculate the statistical indicators of the standard color difference values of all pixel grids in any monitoring area, including the maximum value, the minimum value, the variance and the average value; perform weighted calculation on all indicators in the statistical indicators to obtain the color shadow value of the monitoring area; perform weighted processing on the color shadow values of all monitoring areas and their corresponding area influence weights to obtain the comprehensive color shadow value; Perform weighted processing on the contour area difference, the contour perimeter difference, the similarity and the comprehensive color shadow value to obtain the defect shadow value of the surface image corresponding to the camera; perform weighted processing on the defect shadow values of the surface images corresponding to all cameras to obtain the surface defect evaluation score.

[0013] Preferably, the water level and bubbles in the annular water tank during the defect detection process of the fast food box are visually recognized and analyzed as follows: Use image recognition technology and the preset position of the annular water tank to identify the position of the annular water tank in the surface image; use the edge detection algorithm to locate the edge of the annular water tank and the edge of the water level line, and then calculate the water level height according to the relationship of the image pixel coordinates; set the detection standard duration of the air pump, and mark the time area corresponding to the detection standard duration after the start time of the air pump as the airtightness monitoring time zone; Obtain the water level height at any moment within the airtightness monitoring time zone, and calculate the statistical indicators of the water level height within the airtightness monitoring time zone, including the maximum value, the minimum value, the average value and the variance; Perform weighted processing on all indicators in the statistical indicators of the water level height to obtain the water level influence value; Identify the generation characteristic information of the bubbles in the annular water tank, including the first generation moment, the number, the generation frequency, the bubble size and the rising speed of the bubbles; Set the standard value of any feature in the generation feature information of the bubbles, and subtract the standard value corresponding to any feature in the generation feature information to obtain the standard deviation value; Calculate the statistical indicators of the standard deviation value of any feature in the generation feature information within the airtightness monitoring area, including the maximum value, minimum value, average value, and variance; perform weighted processing on all indicators in the statistical indicators corresponding to the features in the generation feature information to obtain the standard influence value corresponding to the feature; perform weighted processing on the standard influence values of all features in the generation feature information to obtain the bubble influence value; perform weighted processing on the water level influence value and the bubble influence value to obtain the airtightness score.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. The detection table and the annular water tank of the present invention cooperate with each other, and can conveniently seal the periphery of the opening of the fast food box. This sealing method improves the detection effect. The rubber plate and the pressure rod work together to firmly fix the fast food box. The cooperation between the air pump and the air outlet head is wonderful, which can realize the operation of pressurizing the fast food box, and at the same time avoid the residual water stains in the fast food box. The combination of the air pump and the annular water tank provides convenience for detecting the fast food box by gas pressurization and improves the detection efficiency.

[0015] 2. The system of the present invention can accurately capture surface defects from multiple dimensions, and can also deeply detect the airtightness of the fast food box, covering the water level change of the annular water tank and the generation feature information of the bubbles, realizing a comprehensive analysis from the surface to the airtightness, realizing a highly automated detection process, which is efficient and fast, and meets the needs of large-scale production. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] For the convenience of those skilled in the art to understand, the present invention will be further described below with reference to the accompanying drawings.

[0017] Figure 1 It is a schematic diagram of the overall structure of a fast food box packaging defect detection device proposed by the present invention; Figure 2 It is a schematic three-dimensional structure diagram of the cross-section of the pressure rod proposed by the present invention; Figure 3 It is a schematic three-dimensional structure diagram of the overall cross-section proposed by the present invention; Figure 4 For Figure 2 the enlarged schematic diagram of the structure of part A in; Figure 5 It is a schematic block diagram of the principle of a fast food box packaging defect detection system proposed by the present invention.

[0018] Numbers in the figure: 1. Top cover; 2. Box body; 3. Detection table; 4. Annular water tank; 5. Air outlet head; 6. Connecting pipe; 7. Air pump; 8. Filter screen; 9. Connecting seat; 10. Electric push rod; 11. Flap; 12. Spring; 13. Rubber plate; 14. Tipping motor; 15. Connecting rod; 16. Pressing rod; 17. Torsion spring. Detailed implementation manners

[0019] Here, the exemplary embodiments will be described in detail, and the examples are shown in the drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementation manners described in the following exemplary embodiments do not represent all implementation manners consistent with the present disclosure. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0020] Please refer to Figures 1 - 5 As shown, a fast food box packaging defect detection device in the present invention includes a top cover 1. The bottom surface of the top cover 1 is fixedly connected to a box body 2, and a mounting groove is formed in the middle of the top surface of the top cover 1. A sealing ring is provided on the inner wall of the mounting groove, and a detection table 3 is installed inside the mounting groove. An annular water tank 4 is formed in the middle of the top surface of the detection table 3, and a fixing component is installed on the top surface of the detection table 3. The detection table 3 is conveniently installed through the top cover 1; the fast food box is conveniently detected through the detection table 3; the fast food box is conveniently sealed around during detection through the annular water tank 4; an installation opening penetrating to the inner top surface of the box body 2 is formed in the middle of the top surface of the detection table 3, and an air outlet head 5 is installed on the top surface of the detection table 3. The lower end of the air outlet head 5 is placed inside the installation opening, and a connecting pipe 6 is sleeved on the lower end of the air outlet head 5. The gas is conveniently guided through the connecting pipe 6; an air pump 7 is installed inside the box body 2. The air outlet of the air pump 7 is connected to the other end of the connecting pipe 6, and an air inlet is formed in the middle of the rear end surface of the box body 2. A filter screen 8 is installed inside the air inlet. The air pump 7 is convenient for providing gas; the filter screen 8 is used to filter impurities from the external gas inhaled by the air pump 7.

[0021] In the present invention, connection seats 9 are installed at the four ends of the top surface of the detection table 3. Electric push rods 10 penetrate through the connection seats 9. At the far ends of the four electric push rods 10, a retaining piece 11 is fixedly connected. At the near ends of the four electric push rods 10, a rubber plate 13 is fixedly connected. A spring 12 is arranged between the rubber plate 13 and the connection seat 9. The spring 12 is sleeved on the telescopic end of the electric push rod 10. The connection seat 9 is used for installing the electric push rod 10. The telescopic end of the electric push rod 10 is provided with the rubber plate 13. The rubber plate 13 is clamped conveniently through the spring 12. The electric push rod 10 is limited conveniently through the retaining piece 11. Horizontally arranged on the upper end of the top cover 1 is a pressure rod 16. The cross-section of the pressure rod 16 is U-shaped. Vertically fixedly connected to the rear of both ends of the pressure rod 16 are mounting rods. The pressure rod 16 is used for fixing the top of the fast food box. The pressure rod 16 is connected conveniently through the mounting rods. On the far surfaces of the lower ends of the two mounting rods, movable grooves are opened. At the far ends of the two mounting rods, a reversing motor 14 is provided. The reversing motor 14 is installed on the top cover 1. The output end of the reversing motor 14 is connected to a connecting rod 15 arranged in the movable groove. A torsion spring 17 is sleeved on the surface of the connecting rod 15. The other end of the torsion spring 17 is fixedly connected to the inner wall of the movable groove. And the connecting rod 15 is installed on the reversing motor 14. The connecting rod 15 is installed conveniently through the reversing motor 14. The torsion spring 17 is installed conveniently through the connecting rod 15. The mounting rod is reset conveniently through the torsion spring 17.

[0022] It should be noted that the present invention is also provided with a water pump and a water tank used in cooperation with the annular water tank 4. The water pump is arranged at the bottom of the inner cavity of the water tank. The water outlet end of the water pump is connected with a water pipe communicated with the annular water tank. The water tank is used for providing water source for the water pump.

[0023] Working principle: Equipment preparation stage: Connect the air pump 7 with wires and power it on. After the water pump is powered on, it conveys a set amount of water from the water tank to the annular water tank 4. Fast food box placement stage: Control the four electric push rods 10 to move outward, drive the rubber plate 13 to move and compress the spring 12. Start the reversing motor 14 to make the mounting rod drive the pressure rod 16 to lift upward. The torsion spring 17 is in a compressed state. Invert the fast food box and place it on the detection table 3, so that the opening of the fast food box is immersed in water, and the opening of the fast food box is sealed by water. Fast food box fixing stage: Turn off the reversing motor 14. Under the elastic action of the torsion spring 17, the mounting rod drives the pressure rod to press against the top of the fast food box. Turn off the electric push rod 10. Under the elastic action of the spring 12, the rubber plate 13 is pushed to abut against the outer wall of the fast food box. Pressurization detection stage: Start the air pump 7, supply air to the air outlet head 5 through the connecting pipe 6, and then apply air pressure to the fast food box through the air outlet head 5. Result determination stage: If there is a crack in the fast food box, the gas will leak from the crack. If the water level in the annular water tank 4 rises and continuous bubbles emerge in the annular water tank 4, it indicates that the fast food box is intact.

[0024] The present invention also provides a fast food box packaging defect detection system, which includes a vision setting module, an image acquisition module, a defect recognition module, a defect determination module, and a result output module; The vision setting module is used to set a number of high-definition cameras on the detection table 3, and the high-definition cameras are used to collect the surface images of the fast food boxes; among them, the positions of the cameras are reasonably set according to the common shapes, sizes of the fast food boxes and the positions where defects are likely to occur; The image acquisition module is used to collect the surface images of the fast food boxes and send them to the defect recognition module; The defect recognition module is used to receive the surface images of the fast food boxes for visual recognition and analysis to obtain the visual analysis results of the fast food boxes; among them, the visual analysis results include the contour area difference, contour perimeter difference, similarity, color comprehensive shadow value, missing shadow value, and surface defect evaluation score of the surface images; It is also used to perform visual recognition and analysis on the water level and bubbles in the annular water tank 4 during the defect detection process of the fast food box to obtain the airtightness analysis results; among them, the airtightness analysis results include the statistical index of the water level height, the water level influence value, the standard deviation of any feature in the generation characteristic information of the bubbles and the corresponding statistical index, the standard influence value corresponding to the feature, the bubble influence value, and the airtightness score; The defect determination module is used to compare the surface defect evaluation score and the airtightness score with their corresponding preset thresholds. If any value of the surface defect evaluation score or the airtightness score is greater than its preset threshold, it means that the fast food box has defects, and the fast food box is marked as a non-conforming part; on the contrary, if both the surface defect evaluation score and the airtightness score are less than their preset thresholds, it means that the fast food box has no defects, and the fast food box is marked as a conforming part; The result output module is used to receive the visual analysis results and airtightness analysis results sent by the defect recognition module, and store the visual analysis results and airtightness analysis results in the database.

[0025] In the present invention, receiving the surface images of the fast food boxes for visual recognition and analysis specifically includes: Obtain the surface image of the fast food box collected by any camera, process the surface image using an edge detection algorithm to obtain the edge contour of the fast food box, and calculate the area and perimeter of the edge contour respectively to obtain the contour area value and contour perimeter value; obtain the standard contour area and standard contour perimeter of the fast food box collected by the camera; subtract the standard contour area from the contour area value to obtain the contour area difference F1, and subtract the standard contour perimeter from the contour perimeter value to obtain the contour perimeter difference F2; Set the standard edge contour of the fast food box collected by the camera; use the cosine similarity algorithm to calculate the similarity F3 between the edge contour and the standard edge contour; Enlarge the surface image of the fast food box into a pixel grid image, set the standard color value of the fast food box in the pixel grid image corresponding to the camera, identify the color value of any pixel grid in the pixel grid image, and subtract the standard color value of the pixel grid from its color value to obtain the standard color difference value of the pixel grid; Divide the image within the edge contour of the fast food box into several monitoring areas according to the common shape, size and positions where defects are likely to occur, and assign area influence weights to the monitoring areas , where i represents the number of the monitoring area; Calculate the statistical indicators of the standard color difference values of all pixel grids in any monitoring area, including the maximum value FR1, the minimum value FR2, the variance FR3, and the average value FR4; Perform weighted calculation on all the indicators in the statistical indicators to obtain the color shadow value FR of the monitoring area. The formula is expressed as: ; where r1, r2, r3, and r4 respectively represent the weights corresponding to the maximum value, minimum value, variance, and average value in the statistical indicators of the standard color difference value; Perform weighted processing on the color shadow values of all monitoring areas and their corresponding area influence weights to obtain the comprehensive color shadow value F4. The formula is expressed as: ; where N represents the number of monitoring areas divided in the image within the edge contour, and iFR represents the color shadow value of the monitoring area i; Perform weighted processing on the contour area difference, contour perimeter difference, similarity, and comprehensive color shadow value to obtain the defect shadow value F0 of the surface image corresponding to the camera. The formula is expressed as: ; where f1, f2, f3, and f4 respectively represent the weights corresponding to the contour area difference, contour perimeter difference, similarity, and comprehensive color shadow value; Perform weighted processing on the defect shadow values of all surface images corresponding to the cameras to obtain the surface defect evaluation score F. The formula is expressed as: ; where j represents the number corresponding to the camera, n represents the number of cameras, respectively represent the defect shadow value of the surface image corresponding to the camera j and the corresponding weight.

[0026] In the present invention, visual recognition and analysis are performed on the water level and bubbles in the annular water tank 4 during the defect detection process of the fast food box, specifically as follows: Utilize image recognition technology and the preset position of the annular water tank 4 to identify the position of the annular water tank 4 in the surface image; Use the edge detection algorithm to locate the edge of the annular water tank 4 and the edge of the water level line, and then calculate the water level height according to the relationship of the image pixel coordinates; Set the detection standard duration of the air pump 7, and mark the time region corresponding to the detection standard duration after the start time of the air pump 7 as the airtightness monitoring time zone; Obtain the water level height at any moment within the airtightness monitoring time zone, and calculate the statistical indicators of the water level height within the airtightness monitoring time zone, including the maximum value G1, the minimum value G2, the average value G3, and the variance G4; All the statistical indicators of the water level height are weighted to obtain the water level influence value G, which is expressed by the formula: where e1, e2, e3, and e4 respectively represent the weights corresponding to the maximum value, minimum value, average value, and variance in the statistical indicators of the water level height; Identify the generation characteristic information of the bubbles in the annular water tank 4, including the first generation time, quantity, generation frequency, bubble size, and rising speed of the bubbles; Set the standard value of any characteristic in the generation characteristic information of the bubbles, and subtract the standard value from the value of any characteristic in the generation characteristic information to obtain the standard deviation; Calculate the statistical indicators of the standard deviation of any characteristic in the generation characteristic information within the airtightness monitoring area, including the maximum value Y1, minimum value Y2, average value Y3, and variance Y4; All the indicators in the statistical indicators corresponding to the characteristics in the generation characteristic information are weighted to obtain the standard influence value Y corresponding to the characteristics, which is expressed by the formula: where a1, a2, a3, and a4 respectively represent the weights corresponding to the maximum value, minimum value, average value, and variance of the statistical indicators of the standard deviation of any characteristic in the generation characteristic information; The standard influence values of all the characteristics in the generation characteristic information are weighted to obtain the bubble influence value K, which is expressed by the formula: where d represents the number of the characteristics in the generation characteristic information, D represents the number of the characteristics in the generation characteristic information, respectively represent the standard influence value of the characteristic d in the generation characteristic information and the corresponding weight; The water level influence value and the bubble influence value are weighted to obtain the airtightness score P, which is expressed by the formula: where p1 and p2 respectively represent the weights corresponding to the water level influence value and the bubble influence value.

[0027] After considering the specification and practicing the invention disclosed herein, those skilled in the art will readily conceive of other embodiments of the present invention. This application is intended to cover any variations, uses, or adaptations of the present invention, which follow the general principles of the present invention and include known common knowledge or conventional technical means in the technical field not disclosed in this disclosure. The specification and examples are only regarded as exemplary, and the true scope and spirit of the present invention are pointed out by the following claims.

[0028] It should be understood that the present invention is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. A fast food box packaging defect detection device, characterized in that: The invention comprises a top cover (1), characterized in that a box body (2) is fixedly connected to the bottom surface of the top cover (1), a mounting groove is provided in the middle of the top surface of the top cover (1), a sealing ring is provided on the inner wall of the mounting groove, a detection platform (3) is installed inside the mounting groove, an annular water tank (4) is provided in the middle of the top surface of the detection platform (3), and a fixing component is installed on the top surface of the detection platform (3).

2. The fast food box packaging defect detection device according to claim 1 is characterized in that: A mounting opening is provided in the middle of the top surface of the detection platform (3) and extends through the top surface of the box body (2), and an air outlet head (5) is installed on the top surface of the detection platform (3), the lower end of the air outlet head (5) is placed in the mounting opening, and a connecting pipe (6) is sleeved on the lower end of the air outlet head (5).

3. The fast food box packaging defect detection device according to claim 1, characterized in that: An air pump (7) is installed inside the box body (2), the air outlet of the air pump (7) is connected to the other end of the connecting pipe (6), and an air inlet is opened in the middle of the rear end surface of the box body (2), and a filter screen (8) is installed inside the air inlet.

4. The fast food box packaging defect detection device according to claim 1, characterized in that: A connecting seat (9) is installed at the four ends of the top surface of the detection platform (3), and an electric push rod (10) is inserted into the connecting seat (9). The far ends of the four electric push rods (10) are fixedly connected with a blocking piece (11), and the near ends of the four electric push rods (10) are fixedly connected with a rubber plate (13). A spring (12) is provided between the rubber plate (13) and the connecting seat (9), and the spring (12) is sleeved on the telescopic end of the electric push rod (10).

5. The fast food box packaging defect detection device according to claim 1, characterized in that: A pressure rod (16) is horizontally provided at the upper end of the top cover (1); the cross section of the pressure rod (16) is U-shaped, and mounting rods are vertically fixedly connected to the rear of both ends of the pressure rod (16).

6. The fast food box packaging defect detection device according to claim 5, characterized in that: A movable groove is formed at the far end of the lower ends of the two mounting rods, and a flip motor (14) is provided at the far end of the two mounting rods. The flip motor (14) is mounted on the top cover (1), and the output end of the flip motor (14) is connected to one end of a connecting rod (15) provided in the movable groove. A torsion spring (17) is sleeved and connected to the surface of the connecting rod (15), and the other end of the torsion spring (17) is fixedly connected to the inner wall of the movable groove, and the connecting rod (15) is mounted on the flip motor (14).

7. A fast food box packaging defect detection system, using a fast food box packaging defect detection device according to any one of claims 1 to 6, characterized in that: It includes visual setting module, image acquisition module, defect recognition module, defect determination module and result output module; A visual setting module, used to set a plurality of high-definition cameras on the detection platform (3), wherein the high-definition cameras are used to collect surface images of the fast food box; An image acquisition module, used to acquire the surface image of the fast food box and send it to the defect recognition module; The defect recognition module is used to receive the surface image of the fast food box for visual recognition analysis to obtain the visual analysis result of the fast food box; wherein the visual analysis result includes the contour area difference, contour perimeter difference, similarity, color comprehensive shadow value, missing shadow value, and surface defect evaluation score of the surface image; It is also used to perform visual recognition analysis on the water level and bubbles in the annular water tank (4) during the fast food box defect detection process to obtain an airtightness analysis result; wherein the airtightness analysis result includes a statistical index of the water level height, a water level influence value, a standard deviation value of any feature in the bubble generation feature information and the corresponding statistical index, a standard influence value corresponding to the feature, a bubble influence value and an airtightness score; The defect determination module is used to compare the surface defect evaluation score and the air tightness score with their corresponding preset thresholds. If any value of the surface defect evaluation score or the air tightness score is greater than the preset threshold, it means that the fast food box has defects and the fast food box is marked as unqualified. On the contrary, if both the surface defect evaluation score and the air tightness score are less than the preset threshold, it means that the fast food box does not have defects and the fast food box is marked as qualified. The result output module is used to receive the visual analysis results and the airtightness analysis results sent by the defect recognition module, and store the visual analysis results and the airtightness analysis results in a database.

8. A fast food box packaging defect detection system according to claim 7, characterized in that: Receive the surface image of the fast food box for visual recognition analysis, specifically: Obtain a surface image of a fast food box captured by any camera, use an edge detection algorithm to process the surface image to obtain an edge contour of the fast food box, calculate the area and perimeter of the edge contour to obtain a contour face value and a contour perimeter value respectively; obtain a standard contour area and a standard contour perimeter of the fast food box captured by the camera; subtract the standard contour area from the contour face value to obtain a contour area difference, and subtract the standard contour perimeter from the contour perimeter value to obtain a contour perimeter difference; set the camera to capture a standard edge contour of the fast food box; and use a cosine similarity algorithm to calculate the similarity between the edge contour and the standard edge contour; The surface image of the fast food box is magnified into a pixel grid image, the standard color value of the fast food box in the pixel grid image corresponding to the camera is set, the color value of any pixel grid in the pixel grid image is identified, and the standard color difference value of the pixel grid is obtained by subtracting the color value of the pixel grid from its standard color value; the image within the edge contour of the fast food box is divided into several monitoring areas according to the common shape, size and defect-prone position of the fast food box, and the regional influence weight is assigned to the monitoring area; the statistical indicators of the standard color difference values ​​of all pixel grids in any monitoring area are calculated, including the maximum value, minimum value, variance, and average value; all indicators in the statistical indicators are weighted to obtain the color shadow value of the monitoring area; the color shadow values ​​of all monitoring areas are weighted with their corresponding regional influence weights to obtain the color comprehensive shadow value; The contour area difference, contour perimeter difference, similarity, and color comprehensive shadow value are weighted to obtain the shadow missing value of the surface image corresponding to the camera; the shadow missing values ​​of the surface images corresponding to all cameras are weighted to obtain the surface defect assessment score.

9. A fast food box packaging defect detection system according to claim 7, characterized in that: The water level and bubbles in the annular water tank (4) during the fast food box defect detection process are visually identified and analyzed as follows: Using image recognition technology and the preset position of the annular water tank (4), the position of the annular water tank (4) in the surface image is identified; using an edge detection algorithm to locate the edge of the annular water tank (4) and the edge of the water level line, and then calculating the water level height according to the image pixel coordinate relationship; setting the detection standard time of the air pump (7), and marking the time area corresponding to the detection standard time after the start time of the air pump (7) as the air tightness monitoring time area; Obtain the water level height at any time in the air tightness monitoring time zone, and calculate the statistical indicators of the water level height in the air tightness monitoring time zone, including the maximum value, minimum value, average value and variance; All the indicators in the statistical indicators of water level height are weighted to obtain the water level impact value; Identifying the generation characteristic information of bubbles in the annular water tank (4), including the first generation time, quantity, generation frequency, bubble size, and rising speed of the bubbles; Set the standard value of any feature in the generated feature information of the bubble, and subtract the corresponding standard value from the value of any feature in the generated feature information to obtain the standard deviation value; Calculate the statistical indicators of the standard deviation value of any feature in the generated characteristic information within the air tightness monitoring time zone, including the maximum value, minimum value, average value and variance; weight all indicators in the statistical indicators corresponding to the features in the generated characteristic information to obtain the standard influence value corresponding to the feature; weight the standard influence values ​​of all features in the generated characteristic information to obtain the bubble influence value; weight the water level influence value and the bubble influence value to obtain the air tightness score.

Images